Miniworld is a minimalistic 3D interior environment simulator for reinforcement learning & robotics research#

Sequence of observations from Collect Health environment

MiniWorld allows environments to be easily edited like Minigrid meets DM Lab. It can simulate environments with rooms, doors, hallways, and various objects (e.g., office and home environments, mazes).


pip install miniworld


The Gymnasium interface allows to initialize and interact with the Miniworld default environments as follows:

import gymnasium as gym
env = gym.make("MiniWorld-OneRoom-v0")
observation, info = env.reset(seed=42)
for _ in range(1000):
   action = policy(observation)  # User-defined policy function
   observation, reward, terminated, truncated, info = env.step(action)

   if terminated or truncated:
      observation, info = env.reset()